C-RNN-GAN: Continuous recurrent neural networks with adversarial training

TL;DR

This paper introduces C-RNN-GAN, a generative adversarial network designed for continuous sequential data like music, demonstrating improved quality of generated classical music through adversarial training.

Contribution

It presents a novel GAN architecture tailored for continuous sequential data and applies it to generate classical music, showing promising results.

Findings

Generated music improves with training

Statistics indicate high-quality music generation

Generated songs are available for subjective evaluation

Abstract

Generative adversarial networks have been proposed as a way of efficiently training deep generative neural networks. We propose a generative adversarial model that works on continuous sequential data, and apply it by training it on a collection of classical music. We conclude that it generates music that sounds better and better as the model is trained, report statistics on generated music, and let the reader judge the quality by downloading the generated songs.